Computer Simulation of MRSA Transmission in Hospitals

Lay Summary - June 2010

 

Background

Infections due to Methicillin-resistant Staphylococcus aureus (MRSA) are an important cause of morbidity and mortality in hospital patients. Patients can also become colonised with MRSA without symptoms. Colonised patients may have a higher chance of developing an MRSA infection and may transmit MRSA to others.

 

Two main MRSA control methods are patient isolation and decolonisation. These are often combined with screening to detect which patients are colonised. Many screening options exist: some approaches are cheap but take up to four days to give a result, while newer approaches are much more expensive but can give a result in a few hours. There are also questions about who to screen and when. There is no consensus about which strategy to prevent MRSA infection is best, with little available evidence on the effectiveness and value for money of different approaches.

 

Objective

The objective of the study was to assess the effectiveness and value for money of different MRSA infection control strategies in intensive care units (ICUs) and general medical (GM) wards.

 

Methods

We developed a computer simulation model of MRSA transmission within ICUs and GM wards. Individual patients were classified as either susceptible (MRSA negative), colonised or infected. We used the model to examine how the number of infections, deaths and costs changed when different infection control measures were put in place.

 

Preventing MRSA infection contributes to improved quantity and quality of life, summarised by Quality Adjusted Life Years (QALYs) gained. In the UK an intervention is considered good value for money if at least one QALY is gained for each £20,000 - £30,000 spent. In this way we were able to evaluate which interventions were most effective and which represented best value for money.

 

Results and Conclusions

The largest reductions in MRSA infection were achieved by screening combined with decolonisation. In ICUs with low levels of MRSA, such interventions were likely to be good value for money, and could even save money in settings with higher MRSA levels.

 

A policy of decolonisation for all ICU patients was found optimal in the short-term. However, a danger with such widespread use of decolonising antimicrobials is that resistance to them may emerge and spread, making such a policy unsustainable in the long-term. If this risk were considered too great, a policy of rapid screening and decolonisation of those positive was also shown to represent good value for money.

 

Compared to ICUs, the potential for reducing infection by isolation or decolonisation in GM wards was found to be much more limited. All strategies led to much smaller reductions in MRSA infection rates than those seen in ICUs, and screening combined with either decolonisation or isolation was found to be poor value for money.